Data protection systems contain large amounts of data. This data includes personal data, such as financial data, customer/client/patient contact data, audio/visual data, and much more. Corporate computer systems often contain word processing documents, engineering diagrams, spreadsheets, business strategy presentations, and so on. With the proliferation of computer systems and the ease of creating content, the amount of content in an organization has expanded rapidly. Even small offices often have more information stored than any single employee can know about or locate.
Some data protection applications provide functions for actively searching for files within the organization based on a previously created index of the information available in each file. A user can then search for and retrieve documents based on a topic. Typical search software operates on a single index of keywords derived from the data that has been copied for protection purposes. It is typical for an organization to maintain many secondary copies of its data and the various copies are typically stored in multiple formats in multiple devices. For example, when current copy of data is made, previous copies are often maintained so that an historical archive is created. Thus, if the most recent copy does not have the desired data for a restore operation, an older copy may be used. With the existence of multiple copies on multiple devices spanning weeks, months and even years, a search over this data can be complex and time consuming. A search over such a large amount of data can require separately searching content indices of all of the computer systems within an organization. This can put an unexpected load on already burdened systems and can require significant time on the part of a system operator.
Typical search systems also create problems when retrieval of the desired data is attempted. First, typical systems require that retrieval of the identified data be performed as a restore operation. The typical restore operation first identifies a secondary copy of the data in question on a secondary volume and copies the identified copy of the data back onto a production server (or other primary or working volume) and overwrites the existing data files. This can be inconvenient if it is desired to maintain the production copy or if it is merely desired to inspect the contents of a secondary data store. Second, typical systems are blind to the security rights of users and database operators. Typical systems do not have an integrated data rights security control that identifies the security privileges of the operator or user for whom the data is being restored and allows or denies the restore accordingly. Additionally, typical systems do not allow a user to promote and reapply search criteria throughout the data management system.
In the drawings, the same reference numbers and acronyms identify elements or acts with the same or similar functionality for ease of understanding and convenience.